What Trends in Chinese Social Media

Louis Yu
Social Computing Lab HP Labs Palo Alto, California, USA

Sitaram Asur
Social Computing Lab HP Labs Palo Alto, California, USA

Bernardo A. Huberman
Social Computing Lab HP Labs Palo Alto, California, USA

louis.yu@hp.com

sitaram.asur@hp.com

bernardo.huberman@hp.com

ABSTRACT
There has been a tremendous rise in the growth of online social networks all over the world in recent times. While some networks like Twitter and Facebook have been well documented, the popular Chinese microblogging social network Sina Weibo has not been studied. In this work, we examine the key topics that trend on Sina Weibo and contrast them with our observations on Twitter. We find that there is a vast difference in the content shared in China, when compared to a global social network such as Twitter. In China, the trends are created almost entirely due to retweets of media content such as jokes, images and videos, whereas on Twitter, the trends tend to have more to do with current global events and news stories.

Categories and Subject Descriptors
H.4 [Information Systems Applications]: Miscellaneous; I.7.1 [Document and Text Processing]: Document and Text Editing—languages

General Terms
Measurement

Keywords
social network; web structure analysis, China; social computing

1.

INTRODUCTION

Social networks have made tremendous impact on online computing, by providing users opportunities to connect with others and generate enormous content on a daily basis. The enormous user participation in these social networks is reflected in the incessant number of discussions, images, videos, news and conversations that are constantly posted in social sites. Popular networks such as Facebook and Twitter are well-known globally and contain several hundreds of

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. The 5th SNA-KDD Workshop’11 (SNA-KDD’11), August 21, 2011, San Diego CA USA. Copyright 2011 ACM 978-1-4503-0225-8 ...$5.00.

millions of users all over the world. On the other hand, Sina Weibo is a popular microblogging network in China which contains millions of users, almost all of whom are located in China and post in the Chinese language. In China, online social networks have become a major platform for the youth to gather information and to make friends with like-minded individuals [16]. In this regard, a major point of interest is to examine the information that is propagated and the key trend-setters for this medium. There has been a lot of prior research done on the adaptation of influence and evolution of trends in Western online social networks [2] [17] [20]. But, in contrast, Chinese social media has not been well-studied. In this paper, we analyze the evolution of Sina Weibo and provide the first known in-depth study of trending topics on a Chinese online microblogging social network. Our goal is to discover important factors that determine popularity and influence in the context of Chinese social media. To compare, we contrast them with corresponding ones from Western social media (Twitter). We put emphasis on examining how trends are formed and what kind of sources dominate the topics of discussion in Chinese social media. First, we identify and collect the topics that are popular on Sina Weibo over time. For each of these trending topics, we analyze the characteristics of the users and the corresponding tweets that are responsible for creating trends. We believe that this will give us a strong insight into the processes that govern social influence and adoption in China. To perform a comparison, we use similar trending topic data from Twitter. Our key findings are as follows. We observe that there are vast differences between the content that is shared on Sina Weibo when compared to Twitter. In China, people tend to use Sina Weibo to share jokes, images and videos and a significantly large percentage of posts are retweets. The trends that are formed are almost entirely due to the repeated retweets of such media content. This is contrary to what we observe on Twitter, where the trending topics have more to do with current events and the effect of retweets is not as large. We also observe that there are more unverified accounts among the top 100 trend-setters on Sina Weibo than on Twitter and most of the unverified accounts feature discussion forums for user-contributed jokes, images and videos.

2. 2.1

BACKGROUND AND RELATED WORK Degree Distributions

According to the survey by CNNIC in 2010 [9]. retweets and user mentions.1%. In metropolitan cities such as Beijing and Shanghai. In one comprehensive study. movies. Social influence has been studied in a vast array of social networks involving various foci such as interests and personal habits [22] [13] [26].2 Social Influence Studies For many years the structure of various offline social networks has been studied by sociologists (see [15] [23] [6] for surveys). [32] has looked at the adaptation of interests such as books. [8] have performed a comparison of three different measures of influence . Their results show that online social networks follow the power-law in the in-degree and out-degree distributions of user nodes. YouTube. On Twitter.4 The Internet in China 2. 2. teachers in China tend to put aggressive children in a peer group with non-aggressive children. They discovered that traditional media sources are important in causing trends on twitter. they hypothesized that the number of followers may not a good measure of influence. They measured retweets on Twitter and found that passivity was a major factor when it came to forwarding. and Flickr. Douban provides users with review and recommendation services for movies. The first phase was between 1986–1992. In this work. the number of Internet users in China has reached 253 million.indegree. They discovered that the most influential bloggers were not necessarily the most active. an online social network frequently used by the youth in China. Xu et al. The 2010 survey by CNNIC on the Internet development in China [10] reports that the Internet penetration rate in the rural areas of China is on average 5. with Beijing being 46. They also demonstrated with empirical evidence that the number of followers is a poor measure of influence. Tai [31] points out the four major stages of Internet development in China. [11]. the adaptation of influences between editors of Wikipedia articles. Despite this. [30] have looked at the adaptation of aggressive behaviors in a social network of kindergarden children in China. Furthermore. 2. Figure 1: China Age Distribution of Internet Users in The Government plays an important role in fostering the advance of the Internet industry in China. Based on this. the Internet penetration rate has reached over 45%. the number of Internet users in China as of 2010 was reported to be 420 million. and Crandall et al. “with each period reflecting a substantial change not only in technological progress and application. Agarwal et al. This was corroborated by Romero and others [24] who presented a novel influence measure that takes into account the passivity of the audience in the social network. Xin [29] has conducted a survey on BBS’s influence on the University students in China and their behavior on Chinese BBS. books. The development of the Internet industry in China over the past decade has been impressive. [30] have found that over time friendships can be formed between aggressive children and non-agressive children. Recently. [18] examine the linking structure of Flickr and Yahoo!360 and report similar findings. The second phase was between 1992–1995. In contrast. [3] have examined the characteristics of membership closure in LiveJournal. the Chinese Government proposed several large scale network projects and built a national information network infrastructure. Kumar et al.There have been many experiments conducted for studying the structure of online social networks. As a method of controlling aggression. intra-group friendships moderated their aggressive behavior. when Internet applications were limited to the use of emails among a handful of computer research labs in China. It is also the largest online media database and one of the largest online communities in China. Researchers have also analyzed the structure of various Chinese offline social networks [4] [25] [12] [5] [7]. Many of the top retweeted articles that formed trends on Twitter were found to arise from news sources such as the New York Times. music. social influence refers to the concept of people modifying their behavior to bring them closer to the behavior of their friends. Mislove et al.4% and Shanghai being 45. the Internet penetration rate in the urban cities of China is on average 21. [21] [28]. For the aggressive children who are group members.6%. In other work.S.3 Trends on Twitter There are various studies on trends on Twitter [14] [19] .8% [10]. Backstrom et al. music and events. [1] have examined the problem of identifying influential bloggers in the blogosphere. we evaluate how the trending topics in China relate to the news media. Jin [16] has studied the Chinese online Bulletin Board Systems (BBS). [23] have presented a largescale measurement study of online social networks such as Orkut. Xu et al. There are only a few studies of social influence in Chinese online social networks. Cha et al. surpassing the U. events and discussion groups on Douban. and provided observations on the structure and interface of Chinese BBS and the behavioral patterns of its users. In social network analysis. by July 2008. Asur and others [2] have examined the growth and persistence of trending topics on Twitter. but also in the Government’s approach to and apparent perception of the Internet.” 1. 2. According to a survey from the China Internet Network Information Center (CNNIC). Yu et al. the fractional Internet penetration rate in China is still low. taken from [9]). China’s cyberspace is dominated by urban students between the age of 18–30 (see Figure 1 and Figure 2. as the world’s largest Internet market [9].

and 1 “The Internet in China” by the Information Office of the State Council of the People’s Republic of China is available at http://www. the number of followers and followees the user has. We give a brief introduction of Sina Weibo’s interface and functionalities. actively guides people to manage websites in accordance with the law and use the Internet in a wholesome and correct way. we think it is interesting to look at how trends start and evolve in various Chinese online social networks and to analyze the characteristics of trend-setters determining if they represent Government organizations. The new applications and services on the Internet have provided a broader scope for people to express their opinions. the Chinese Government blocked the access to Twitter and Fanfou. Similar to Twitter. and social networking platforms to engage in activities such as exchanging viewpoints and sharing information [16]. commercial organizations.sina. in August 2009. the media.cn/i/2010-1116/10314870771. etc. blogs. the Weibo microblog now has more than 50 million active users per day. According to The Internet in China 1 released by the Information Office of the State Council of China: The Chinese government attaches great importance to protecting the safe flow of Internet information. video sharing and social networking websites are developing rapidly in China and provide greater convenience for Chinese citizens to communicate online. the then leading Twitter clone. 3.Figure 2: The Occupation Distribution of Internet Users in China 3. there are over a million BBSs and some 220 million bloggers. Sina has reported in the annual report that it has more than 60.htm 2 A netizen is a person actively involved in online communities [27].5 Chinese Online Social Networks Online social networks are a major part of the Chinese Internet culture [16]. during which time the Internet has become a powerful medium in the Chinese society.scio. A user profile also displays the user’s recent tweets and retweets. and to fully express their opinions and represent their interests. microblog. a brief description of the user. discussion groups. According to The Internet in China: Vigorous online ideas exchange is a major characteristic of China’s Internet development. sports stars. According to a sample survey. SINA WEIBO From the above motivation. each day people post over three million messages via BBS. Actively participating in online information communication and content creation. On July 2009.shtml 3 . or individuals. there are some differences in terms of the functionalities offered.. Internet companies such as Sina and Tencent started offering microblog services to their users in mainland China. China’s biggest web portal. 4.000 verified accounts consisting of celebrities.com. including blog. the Government started to implement a variety of technological and policy control mechanisms to regulate the safe flow of the information on the Internet. The Chinese Government stepped up its effort in building the information network infrastructure. and 10 million newly registered users per month. the huge quantity of BBS posts and blog articles is far beyond that of any other country. While both Twitter and Sina Weibo enable users to post messages of up to 140 characters. there are two types of user accounts on Sina Weibo. and over 66% of Chinese netizens frequently place postings to discuss various topics. According to the Sina corporation annual report 3 . hoping that the IT industry would yield significant benefits to the nation’s economy. The third phase was between 1995–1997. and the number of tweets the user made. Sina Weibo was launched by the Sina corporation. The fourth phase started from 1998 and continues to the present. A user profile on Sina Weibo displays the user’s name. news commentary sites. in China. We choose to analyze the characteristics of trends and trend-setters on Sina Weibo.cn/zxbd/wz/201006/t667385. The newly emerging online services. well known organizations (both Government and commercial) and other The Sina corporation annual report 2011 is available (in Chinese) at http://tech.gov. 3. In China. with over 80% of them providing electronic bulletin service.1 User Profiles 2. blogs. A verified user account typically represents a well known public figure or organization in China. Meanwhile. netizens have greatly enriched Internet information and content. Netizens2 in China organize themselves using forums. China’s websites attach great importance to providing netizens with opinion expression services. regular user accounts and verified user accounts.

it can only be accessed under the original message. which also presents a constantly updated list of trending topics. The first is an original tweet made by a user. Figure 3 illustrates some messages with embedded pictures and videos on Sina Weibo. Figure 5: The List of Hourly Trending Keywords (with Translations) We monitored the list of hourly trending keywords every hour for 30 days and retrieved every new keywords appeared in the list. Figure 3: An Example of Embedded Videos and Pictures (Translations of the Tweets Omitted) 3. pictures. which are keywords that are most frequently used in tweets over that period. The equivalent of a retweet on Sina Weibo is instead shown as two two amalgamated entries: the original entry and the current user’s actual entry which is a commentary on the original entry (see Figure 4). Sina Weibo also has a functionality absent from Twitter: the comment. Figure 6 a) illustrates . we can see that the original message is retweeted 62 times and commented 10 times by other users. They are ranked according to the frequency of appearances. To compare with Twitter. Figure 4: An Example of Comments and Retweets (Translations of the Tweets Omitted) illustrates the list of hourly trending keywords (with translations). While Twitter users can post tweets consisting of text and links.VIPs. or rebroadcasting someone else’s messages to one’s followers. This is similar to Twitter. we calculated the distributions for the number of tweets and topics in our dataset. we can see that the retweeting and commenting buttons are listed under the tweet.32 million tweets on 3361 different trending topics over 40 days using the Twitter Search API. Sina Weibo users can post messages containing text. Figure 4 illustrates two example tweets on Sina Weibo. EXPERIMENTS AND RESULTS First. When a Weibo user makes a comment. videos and links. A common practice on Twitter is “retweeting”. we obtained 16. We retrieved in total 4411 new trending keywords over the 30 days observation period. Figure 5 4.3 Retweets and Comments Twitter users can address tweets to other users and can mention others in their tweets [13]. 3. The second is a retweet.2 The Content of Tweets on Sina Weibo There is an important difference in the content of tweets between Sina Weibo and Twitter. it is not rebroadcasted to the user’s followers. 3.4 Trending keywords Sina Weibo offers a list of 50 keywords that appeared most frequently in users’ tweets over the past hour. Instead.

87 3. stories and so on. an online travel magazine. they forward it to their followers. other followers tend to retweet them frequently.32 25286. We see from the description of the user that this account welcomes submission from followers.92 35. In this case. The corresponding most retweeted users for trending topics on Twitter is shown in Table 2. Table 2: Top Retweeted Users on Twitter contributing to at least 50 trending topics each 4. When we further inspected these accounts.08 Retweet-Ratio 179.94 23579.82 31695. the list is . Table 1 illustrates the top 20 most retweeted authors appearing in at least 10 trending topics each. From Figure 7 we also see a contribution from a follower.58 16.48 46132.81 30310.5 25576.52 100.88 32 31.43 35. For every new trending keyword we retrieved the most retweeted tweets in the past hour and compiled a list of most retweeted users.17 46. We further inspect one of the accounts: ID 1644395354 (see Figure 7).57 33. a brief translation of the description of the authors (author description). When people find a tweet interesting either due to the content or the source. the number of topics the authors’ tweets appeared in (topics). We define an author’s retweet ratio as the number of times the authors’ tweets are retweeted divided by the number of trending topics these tweets appeared in. For each author we have included the ID of the user account (ID). both these distributions follow the power law. the number of times these tweets are retweeted.06 62.25 65338.76 5.19 84.81 100. Figure 6 b) illustrates the distribution for the number of users (Y-axis) whose tweets appear in numbers of topics (X-axis). movie trivia. and a Chinese celebrity.12 28460 26918. the influential authors are ranked according to their retweet ratios.39 the distributions for the number of users (Y-axies) with a certain number of tweets (X-axis) in our list of trending topics.13 39631. 4.78 15. The users who follow these accounts tend to contribute jokes or stories. Once they are posted.48 14.10 9.81 84.21 24.23 39420.67 23807.1. we discovered that these accounts seem to operate as discussion and sharing platforms.17 40683. and finally.52 3.35 50.60 19. As we can observe from the figure.1 Trend-setters on Sina Weibo One of the main forms of information propagation in social networks such as Twitter and Sina is through retweets. a fashion brand. They all seem to have a strong focus on collecting usercontributed jokes. Followers of this account can email jokes to the account and the account administrator will post them. whether it is a verified account. quizzes. the number of tweets the author made in the trending topics (tweets).1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ID 1757128873 1643830957 1670645393 1992523932 1735618041 1644395354 1843443790 1644572034 1674242970 1713926427 1657430300 1195230310 1750903687 1757353251 1644570320 1802393212 1920061532 1644574352 1780417033 1644394154 Table 1: Top 20 Retweeted Users in At Least 10 Trending Topics Author Description (Translated) Verified Account Retweets Tweets Topics Urban Fashion Magazine Yes 1194999 37 12 Fashion Brand VANCL Yes 849404 21 13 Online Travel Magazine Yes 127737 123 21 Gourmet Factory No 553586 86 12 Horoscopes No 1545955 101 38 Silly Jokes No 3210130 258 81 Good Movies No 1497968 140 38 Wonderful Quotes No 602528 39 17 Global Music No 697308 116 22 Funny Jokes Countdown No 3667566 438 121 Creative Ideas No 742178 111 25 Famous Chinese singer Yes 284600 25 10 Good Music No 323022 52 12 Movie Factory No 1509003 230 59 Strange Stories No 1668910 250 66 Beautiful Pictures No 435312 33 18 Global Music No 432444 65 18 Female Fashion No 809440 87 34 Useful Tips No 735070 153 31 Funny Quizzes No 589477 77 25 Author vovo panico cnnbrk keshasuja LadyGonga BreakingNews MLB nytimes HerbertFromFG espn globovision huffingtonpost skynewsbreak el pais stcom la patilla reuters WashingtonPost bbcworld CBSnews TelegraphNews tweetmeme nydailynews Retweets 11688 8444 5110 4580 8406 3866 2960 2693 2371 2668 2135 1664 1623 1255 1273 957 929 832 547 464 342 173 Topics 65 84 51 54 100 62 59 58 66 75 63 52 52 51 65 57 60 59 56 79 97 51 Retweet-Ratio 99583.21 35442.06 23711.1 Authors From Table 1 we observed that only 4 out of the top 20 influential authors were verified accounts. The other 16 influential authors are unverified accounts. This account focuses on posting tweets about jokes.77 57987.46 29687. The 4 verified accounts represent an urban fashion magazine.52 24184 24024.

Figure 6: The Distribution for the Number of Tweets and the Number of Topics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Table 3: Profile Information for Top 20 Retweeted Users Images(%) Videos(%) Links(%) Followees Followers Tweets 70% 0% 32% 673 461398 719 57% 71% 0% 715 300358 2508 21% 0% 17% 67 597063 2600 30% 0% 0% 20 245026 518 20% 0% 0% 81 1884896 4261 8% 0% 0% 650 3536888 10598 15% 11% 0% 12 625117 804 46% 0% 0% 368 2338610 3605 0% 22% 0% 79 405847 716 5% 1% 0% 11 2411888 17818 14% 4% 0% 634 1551438 4899 44% 16% 0% 352 6107858 1850 0% 50% 0% 1136 590099 2041 6% 7% 0% 303 1210833 11411 10% 0% 1% 555 1220027 4249 45% 0% 0% 13 615461 1254 5% 40% 0% 12 496171 571 30% 0% 0% 60 901612 3506 15% 3% 0% 9 763264 2718 25% 0% 0% 4 853877 2362 .

The T weets column in Table 1 gives the unique tweets that have been retweeted. jokes. The overall retweet percentage was around 62% for the trending topics. we also examine the content of their tweets that appeared in the trending topics. and the total number tweets they have made since their accounts are activated (Table 3). Videos and Links in Tweets dominated by popular news sources such as CNN.6% of the tweets on trending topics. We observed the verified accounts are of celebrities. and links. This implies that the topics are trending mainly because of some content that has been retweeted many times.1. the New York Times and ESPN. magazines and a few other media sources. we found this to be true on Twitter as well (shown in Table 4). We observe that a large percentage of the tweets of these users include an embedded image. This represents an important contrast in the use of these media.5 Verified Accounts When we considered the top 100 trend-setters. For example. we looked at the number of followers and followees they have. On the other hand. videos. We hypothesize that these influential authors do not actively seek out accounts to follow. Instead they are unverified accounts acting as discussion forums and a platform for users to share funny pictures. Interestingly. and many tweets included an embedded video or a link. On the other hand. newspapers. For the list of trend-setters on Sina Weibo (Table 1).2 Retweets When we consider the ratio of retweets. although they contribute to fewer topics. 4. Table 3 illustrates the percentage of these users’ tweets which included images. We observe that none of the unverified accounts in Table 1 are personal accounts. with Chinese users being more inclined to share and propagate trivial content than the Twitter users. and stories.1.Author vovo panico cnnbrk keshasuja LadyGonga BreakingNews MLB nytimes HerbertFromFG espn globovision huffingtonpost skynewsbreak el pais stcom la patilla reuters WashingtonPost bbcworld CBSnews TelegraphNews Followees 1069 41 0 37 382 18829 465 763 286 3582 4684 5 46226 12 51 603 284 20 122 238 Followers 154589 4380908 88 136433 2570662 1237615 3250977 23318 1326168 753440 1042330 198349 572260 59763 306965 724204 458721 796009 1716649 38599 Table 4: Follower/Followee relationships for Top Retweeted Twitter Users trending topics. The number of retweets that authors get on Sina Weibo are several orders of magnitude greater than the retweets for the trending topics on Twitter. for Twitter trends.3 Embedded Images. with the top retweeted users having very skewed follower/followee ratios. Figure 7: An Illustration of an User Account on Sina Weibo 4. we observe that the Sina Weibo user accounts whose tweets are retweeted frequently and appear in multiple trending topics over time are mostly estab- . This is again demonstrative of the type of content that is shared in these two social media services. their effect is not so large.2 Random Profile Analysis From the above analyses.1. In contrast. 4. 4. A large percentage of the topics that trended accordingly dealt with events in the news. We can observe that the rate at which they have been retweeted is phenomenal. we found that only 23 were verified accounts. While retweets do contribute to making a topic trend on Twitter. we once again observe a strong contrast with Twitter. It is their content which attracts other users to follow them. Twitter users post links in only 17. the retweets form only 31% of the overall tweets for the For each of the influential authors.1. the consistent trend-setters on Sina Weibo are not media organizations. The descriptions of these 23 are shown in Table 5. This indicates that Twitter users are more attuned to news events than Sina Weibo users and amplify such articles through the medium of Twitter. We discovered that most of the influential authors have more followers than followees. the top retweeted user posted 37 tweets which were totally retweeted 1194999 times. with only 1 out of 4 verified accounts (in the top 20) belonging to a media organization.4 Follower Relationships 4.

We see that the distribution for the number of users peak at around 55% and 80% while remain low below 40% and above 85%. To verify the above hypothesis.Figure 8: The Distribution for the Percentage of Tweets that are a) Retweets and b) include images Figure 9: The Distribution for the Percentage of Tweets that include a) Videos and b) Links lished for discussion and image/video sharing purposes. Followers use these accounts to share interesting stories. and over half of the tweets include an embedded image. links and the percentage of tweets that are retweets.03% of the tweets include a link (91. It has been shown in [24] that URLs form 1/15th (6. In the case of twitter. and pictures. For each user. we selected 1732 random users on Sina Weibo. 60%. videos.24% of the tweets are retweets (49. all forms of media are shared through the use of hyperlinks. Content: Figure 8 b) illustrates the histogram for the number of users with a certain percentage of tweets which include an embedded image.6%) of all twitter .43% of the tweets include an embedded image (43. jokes. 56. only 5. we retrieved his/her last 100 tweets and analyzed the percentage of the tweets which include images. We see that both distributions peaked at 5% and then diminish quickly after.43% did not) and 8. We hypothesize that in general.57% of the tweets include an embedded video (94. Figure 8 a) illustrates the histogram for the number of users with a certain percentage of tweets which are retweets.57% of the tweets do not). Figure 9 a) and 9 b) illustrates the histograms for the number of users with a certain percentage of tweets which include a video or a link.76% of the tweets are original tweets). 70% and 80%. We see that the distributions for the number of users are fairly even throughout all the percentages and is especially high around 45%. We also hypothesize that a high percentage of these users’ tweets tend to include an image or a video for illustration purposes. users of Sina Weibo tend to retweet information more often than Twitter users. We observe that over half of the tweets by our 1732 random users are retweets. Retweets: We find that on average 50.94% did not).

[20] J. Tomkins. May 2010. [4] Y. [9] CNNIC. [16] L. The trends that are formed are almost entirely due to the repeated retweets of such media content. Jin. and B. Asur. 9(4):471–488. Kwak. The influence of relational demography and guanxi: the Chinese case. In Proceedings of the 12th International Conference on Knowledge Discovery and Data Mining. 2011. class. 2005. Galaskiewicz. We also plan to identify the patterns of retweets between followers of influential authors. 5. . G. [19] H. [5] Y. J. Languages and Programming. WSDM’08. P. we plan to conduct further temporal analysis regarding the evolution of trends on Sina Weibo. 2010. Influential nodes in a diffusion model for social networks. Crandall. Abolhassani. Benevenuto. CONCLUSION AND FUTURE WORK We analyzed the tweets that are responsible for causing trending topics on Sina Weibo and the users that created these tweets. 62(3):366–385. Trends in social media . ACM. 6.-S. Feedback effects between similarity and social influence in online communities. Tardos. Social networks that matter: Twitter under the microscope. Huberman. Kempe. WWW ’10. and J. Leskovec. Liu. [3] L. [8] M. Kleinberg. 2010.persistence and decay. [6] M. Group formation in large social networks: membership. [7] P. pages 380–389. [15] M. A. 2010. J. pages 160–168. We observed that there are vast differences between the content that is shared on Sina Weibo than that of Twitter. pages 66–72. and S. Tsui. Survey report on internet development in rural china (in chinese). Kleinberg. ACM. and X. D. we observed on Twitter that trending topics are mainly caused by sources of media [2]. Goyal. [14] B. Breiger. Organization Science. and evolution. images and videos and a significantly large percentage of posts are retweets. Tang. In Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining. Massachusetts Institute of Technology. 83(4):1443–1468. Identifying the Influential Bloggers in a Community. C. Bian. J. In Proceedings of 32nd International Colloquium on Automata. Bonchi. Wu. and L. Wasserman. and E. Farh. Models and Methods in Social Network Analysis. 2006. D. [11] D. pages 1127–1138. Moon. pages 591–600. A. Lee. In Proceedings of Pacific-Asia Conference on Knowledge Discovery and Data Mining. Different aspects of social network analysis. 1998. Lakshmanan. W. F. Romero. Cosley. D. Measuring User Influence in Twitter: The Million Follower Fallacy. A. 2005. editors. Huttenlocher. Structure and evolution of online social networks. Agarwal. S. In Web Search and Data Mining. Patterns of influence in a recommendation network. and A. growth. 2006. People tend to use Sina Weibo to share jokes. Xin. Suri. 2008. Gummadi. network bridges. and K. Lan. May 2003. Jamali and H. A. and J. In 5th International AAAI Conference on Weblogs and Social Media. L. a social network or a news media? In Proceedings of the 19th international conference on World wide web. In Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. K. Norton & Company. This is a very low percentage when compared to Sina Weibo. Kleinberg. REFERENCES [1] N. ACM.-L. pages 241–250. Wang. and C. The 21st statistics report on the internet development in china (in chinese). Cheng. As future work. 2006. In Proceedings of the 2006 IEEE/WIC/ACM International Conference on Web Intelligence. Kumar. Occupation. [13] A. 2005. Cha. [2] S. pages 44–54. J. Learning influence probabilities in social networks. J. Carrington. Nexus: Small Worlds and the Groundbreaking Theory of Networks. B. April 2009. S. Springer-Verlag. [12] J. Szabo. Park. Singh. H. [18] R. and S. [17] D. Huttenlocher. 2008. H. Novak. American Sociological Review. and S. and P. 1997. Yu. Chinese outline BBS sphere: what BBS has brought to China. D. Scott. Haddadi. In contrast. What is twitter. Buchanan. D. Kleinberg. 2008. Bringing strong ties back in: indirect ties. R. V. and social networks in urban china. W. In Fourth International AAAI Conference on Weblogs and Social Media. and job searches in china. Social Forces. Springer Verlag. M. [10] CNNIC. J. S. Cambrige University Press. H. 2010. Computing Research Repository. Master’s thesis. and F. F. pages 611–617. Backstrom. Huberman. Davis.Rank 1 2 2 12 21 25 26 42 43 46 48 58 62 70 71 74 75 82 83 84 95 96 100 ID 1757128873 1643830957 1670645393 1195230310 1740006601 1730380283 1760945071 1322920531 1771665380 1266321801 1883881851 1698229264 1642591402 1743374541 1618051664 1653689003 1640601392 1195031270 1835254597 1830442653 1765148101 1258256457 1596329427 Description Fashion Web Magazine Fashion Brand Travel Web Magazine Celebrity Celebrity Game Discussion Forum Chinese Groupon Celebrity Record Label Celebrity Organization (NBA China) Music Web Magazine Sina Entertainment Pictures Discussion Forum Sina News Newspaper Sina Video Celebrity Music Web Magazine Music Web Magazine Sina Fashion Celebrity Celebrity Table 5: Verified Accounts Among Top 100 Trendsetters traffic. Bian. 2005.

W. . M. In Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining. and Z. 4:100–104. Interpersonal networks and workplace controls in urban china. J. Social networks and aggressive behavior in chinese children. J. M. McPherson. In 20th International World Wide Web Conference (WWW’11). 29:89–105. Druschel. M. Routledge. 27(1):415–444. Measurement and analysis of online social networks. A. Mathioudakis and N. Y. A. and J. Galuba. Hofman. pages 29–42. International Journal of Behavioral Development. Who says what to whom on twitter. The evolution of friendships in chinese online social networks.0: where should we go? IEEE Intelligent Systems. ACM. Security. and L. King. Huberman. D. KDD ’09. Chang. J. Watts. Chinese bulletin board system’s influence upon university students and ways to cope with it (in chinese). [28] S. 2010 IEEE International Conference on. Tang. A. P. 2004. 2003. The Internet in China : Cyberspace and Civil Society. Schwartz. Annual Review of Sociology. [29] M. 2011. Social influence analysis in large-scale networks. Influence and passivity in social media. 2006. Journal of Nanjing University of Technology (Social Science Edition). In Proceedings of the 7th SIGCOMM Conference on Internet Measurement. and B. In Proceedings of the 20th international conference on World wide web.[21] M. Wu. [24] D. 24(3):2–4. 0:81–87. 2009. 28:401–410. K. 1993. Cook. and D. Ruan. Xue. Mislove. and B. Risk and Trust. C. Xu. Bhattacharjee. Wang. M. pages 705–714. 2009. 2010. [26] J. Romero. Z. The Australian Journal of Chinese Affairs. Social Computing / IEEE International Conference on Privacy. Beyond x 2. SIGMOD ’10. [23] A. [25] D. Sun. 2007. [30] Y. 2001. L. Xin. W. 2011. Birds of a feather: homophily in social networks. Koudas. [27] F. Asur. Farver. [32] L. Twittermonitor: trend detection over the twitter stream. Mason. M. Yang. S. [31] T. WWW ’11. pages 807–816. Gummadi. Wang. P. Yu and V. pages 1155–1158. 2010. [22] M. In Proceedings of the 2010 international conference on Management of data. Smith-Lovin. Marcon. J.

Sign up to vote on this title
UsefulNot useful